Jump directly to the Content
Jump directly to the content

J. Bottum

The Subjunctive That Killed Hugh Finn

Our language about what a patient "would" want turns sympathy into empathy, pity into murder.

On March 9, 1995, a car driven by television broadcaster Hugh Finn was smashed in a head-on collision on an icy road, leaving him with serious neurological damage. Though some physical examinations—and the testimony of those who spent enough time with him to notice small changes—suggested that he alternated unconscious sleep cycles with regular conscious periods, he lay unmoving for the next three years.

Early last summer, accepting medical advice that Finn would never improve, his wife, Michele, ordered his life ended. And after months of fighting off legal challenges from her husband's family, she at last succeeded. On October 9, 1998, Hugh Finn died of starvation and dehydration in a nursing home in Manassas, Virginia, eight days after doctors removed the tubes providing him food and water.

The twists and turns of the Finn case were heavily reported in the local Washington, D.C., and northern Virginia press. And almost every newspaper and television account contained the same quotation, a line in which Hugh Finn's wife put forward her fundamental argument for stopping his care. Sometimes it was given as Michele Finn's indirect report of a conversation with her husband in 1993: "He said he would never want to live like that"; sometimes it was stripped down to her intuition of his nature: "He wouldn't want to live like that." But, particularly in the shorter form, it was exactly what was so strenuously denied by the family members whose dispute with Finn's wife made the case newsworthy, and to live in Washington was to read or hear the line nearly every day from June to October.

There's nothing new about such claims to know the desires of those who cannot speak for themselves. The anti-euthanasia lawyer and author Wesley J. Smith has argued that Finn's condition received notice only because it involved a fairly prominent figure with relatives able to publicize their objections. The Virginia court's ruling that allowed the removal of Hugh Finn's food and water is typical of the way such cases are routinely decided. And Michele Finn's argument for ending her husband's life is typical of the kind of proof routinely taken as convincing.

In fact, this sort of claim - -a description of what would be wanted by someone presumed incapable of having or expressing wants—is ubiquitous nowadays. It is the euthanasia advocates' most intuitively persuasive slogan, recognized as compelling by the courts. It is heard on the lowest-brow talk shows and read in the highest-brow publications. In the fall of 1996, six of America's most famous philosophers—Ronald Dworkin, Thomas Nagel, Robert Nozick, John Rawls, Thomas Scanlon, and Judith Jarvis Thomson—submitted to the Supreme Court a brief arguing that the legalization of assisted suicide is necessary to alleviate "the anguish some would feel at remaining alive, but intubated, helpless, and often sedated near oblivion."

But it was only under the constant bombardment of Washington news reports about Hugh Finn that the real strangeness of the claim first struck me. The notion that "he wouldn't want to live like that" initially seems both plausible and humane, but closely examined it becomes increasingly confusing: an impulse of pity somehow transformed into a demand that we be pitiless and hard.

If one stops for a moment to consider it, there is something very odd about the "would" that invariably occurs in such constructions. It is, in the technical language of grammar, is a "modal auxiliary," indicating one of the few remaining common uses of the subjunctive mood in current American English. But why are such claims phrased that way in the first place? Why, if Mrs. Finn wants to convince us that she knows her husband's desire, does she naturally fall back on the subjunctive present, "He wouldn't want to live like that" rather than the straightforward indicative present, "He doesn't want to live like that," or even the indicative past, "He didn't want to live like that"?

Part of the answer appears with grammatical parsing. Taken by itself, Michele Finn's claim is incomplete. Diagrammed in the way schoolmarms used to demand, it emerges as the main clause of a sentence with a strongly implied adverbial clause, a contrary-to-fact condition that modifies the verb and forces it into the subjunctive. We understand the claim because we intuitively grasp that it assumes some unspoken but discernible conditional: He wouldn't want to live like that if he understood his present condition, or if he could see how he looks, or if he were still capable of having a normal person's wants, or if he were the man he used to be.

Parsed like this, however, the claim starts to lose some of its attractive glow. And it dims even more when we realize that the main clause is intended as an argument for agreeing to his subjunctive, conditional wish to have us stop giving him food and water. Spelled out in its entirety—and rendered as neutrally as possible—Michele Finn's argument that her husband "wouldn't want to live like that" actually means something like this:

If Hugh Finn fully understood his helpless, silenced, and undignified condition, and if he could express his rational desires, then he would ask us to end his life—and since we ought to grant the reasonable requests of those under our care, we should therefore remove his feeding tube.

There is a tautological sheen to this proposition that encourages us to overlook the strong likelihood that Finn didn't fully understand his condition, and the certainty that he wasn't able to express his desires regarding his wife's intention to cut off his food and water. And that deceptive quality derives ultimately—if more subtly than a dogmatic logician might admit—from the defects of its logical structure.

In everyday speech, we often make statements in the form "If a certain antecedent condition is met, then a certain consequent state of affairs will result": "If Michael Jordan comes out of retirement yet again, then the Chicago Bulls will win another NBA championship"; "If the floodwaters rise three more feet, then the dam will break." And we typically intend these statements to represent miniature arguments expressing some sort of causal relation between the "if" clause and the "then" clause, the antecedent and the consequent.

When logicians examine such conditionals, however, they tend to consider them not as arguments but as well-formed propositions, complete in themselves. And, judged by this standard, all propositions with the form "If an antecedent condition, then a consequent result" turn out to be true when the antecedent is false, regardless of what is asserted by the consequent. "If the moon is made of green cheese, then pigs can fly" is a perfectly true proposition, and so is "If the moon is made of green cheese, then pigs cannot fly."

This outcome is obviously counterintuitive. We want such propositions to be meaningful: perhaps true, perhaps false, but certainly open to reasonable argument. We want "If General Cornwallis had defeated George Washington in the Revolutionary War, then present-day America would resemble Canada" to be worth discussing, precisely because we expect there to be a connection between the antecedent and the consequent: "Yes, it's false that Cornwallis defeated Washington, but if he had, there would be definite and imaginable consequences down to the present day."

But on the highest level of abstraction, in the pure logic of such propositions considered formally, a false antecedent gives birth with perfect consistency to any consequent whatsoever. If the moon is made of green cheese, then—well, then, anything goes. As long as the moon is not made of green cheese—as long as the antecedent is false—it doesn't matter what the consequent says; the proposition "If an antecedent condition, then a consequent result" remains true. It is true, however, only in the way that tautologies are true, only because all hypothetical propositions with a false antecedent are true: uninformatively true, meaninglessly true, and providing no clue about the truth of the consequent.

At first hearing, a proposition like "If Hugh Finn knew his present condition, then he would want us to end his life" sounds as though it were asserting a causal connection rather than merely a logical one—as though it resembled "If Michael Jordan comes out of retirement yet again, then the Chicago Bulls will win another NBA championship" rather than "If the moon is made of green cheese, then pigs can fly."

But a logician might observe that, whatever the claimed causal connection in the Finn proposition, deciding on the basis of it to end Finn's life severely deepens the logical problem, for killing the man means doing the one thing that renders the antecedent forever false. Maybe he'd want us to remove his feeding tube and maybe he wouldn't, but making him dead eliminates even the thin possibility that he might ever know his condition—which is ostensibly what allowed us to decide to end his life in the first place.

In point of fact, we would probably have to hunt a long way to find a logician to give this explanation, someone willing to take the grieving Mrs. Finn aside and explain to her that, really, she ought to watch what she says, since conditionals with false antecedents are tautological. That there are serious philosophical problems involved in the end of life, no one denies. But we haven't had a great deal of help from America's philosophical community in sorting them out. Many of these problems are grammatical and logical confusions of the very kind that philosophers are trained to analyze, and what we have witnessed in recent euthanasia debates is an abdication of professional responsibility: a refusal by philosophers (with a few exceptions) to help purge public discourse of bad arguments.

Perhaps this failure derives from a desire to say something adequate to a terrible occasion, or perhaps from a willingness to subordinate philosophical expertise to obtaining a desired social-policy goal, or perhaps simply from a residue of good manners that shies from lecturing the grief-stricken caretakers of coma victims and the terminally ill. Over the past 20 years, as the fields of critical legal theory and bioethics have boomed, an enormous number of philosophers have made public interventions in the lawyerly technicalities of constitutional jurisprudence and the practical ethics of medicine—areas in which there is, in fact, little proof of the helpfulness of high-level philosophical analysis. What we haven't received is much public presentation of the logical explication that philosophers are uniquely qualified to provide.

So, for instance, when Dworkin, Nagel, Nozick, Rawls, Scanlon, and Thomson filed with the Supreme Court their assisted-suicide brief (subsequently published in the March 27, 1997, issue of the New York Review of Books), they spoke three times of acting in the "best interests" of patients by assisting in their suicides—without ever doing the one thing for which we genuinely need trained philosophical aid: providing an explanation of how it can ever be in anyone's best interest to cease to have interests.

The possibility of a person's best interests being met by dying—as something phrased in everyday language or even in technical legal terms—is a common argument for euthanasia. One has heard it many times before, and it has a certain surface plausibility that may be aiming at a significant thought. My own guess is that the notion of being well served by death will prove, once fully spelled out, to require the religiously significant assumption of an after-death vantage point from which to view the whole of the now-completed life. But in any case, the argument poses logical difficulties, and its real significance can emerge only with the considerable unpacking that "The Philosophers' Brief" refused to perform.

The six American philosophers who wrote the brief may have feared that the justices on the Supreme Court (and the readers of the New York Review of Books) wouldn't be able to follow the technicalities of a purely philosophical analysis. But when, for example, they write, "Death is, for each of us, among the most significant events of life," they fall to something that would be astonishingly subphilosophical even coming from someone who had never studied their discipline. Surely philosophers are the ones we expect to take account of the fact that, for the dying person, death is not an event in any usual sense of the word. And even if there were some way to imagine it as an event, philosophers are supposed to be the ones who realize that death certainly can't be an event of life. (Wittgenstein, in fact, has a famous passage making this point.) And even if we allow it to stand as an event of life, philosophers—from Socrates on—are the ones we want to remind us that death is not merely "among the most significant events of life" but ranks all by itself.

There are real costs to this kind of abdication of philosophical responsibility—this refusal, while making public arguments, to make public the confusions present in the common arguments about euthanasia. And a clear example of what we might gain, were philosophers to help us sort out what is being asserted in everyday speech, comes with a close examination of Michele Finn's argument that her husband "wouldn't want to live like that."

We needn't accept the results of logical analysis in their entirety to gain from them a sense that there is something disturbing in the use of the Finn proposition as an argument for ending lives. An awareness of the abstract problem with conditionals can make us uneasy with Mrs. Finn's description of her husband—can begin to explain why Finn's other family members felt themselves to be struggling through a fog of claims they were certain were wrong but could find no clear method to answer. ("John Finn believes Hugh Finn would not want the feeding tube removed if he were able to express his wishes," declared his brother's losing petition to have Michele Finn replaced as legal guardian.)

But there is an even more important way in which philosophical analysis aims us toward understanding what is actually going on when someone claims to know what another person would have wanted.

Logic suggests that the problems of the Finn proposition are deepened when we act on its apparent conclusion—when we end Hugh Finn's life and thereby eliminate even the remotest possibility that he would ever understand his condition well enough to request that we go ahead and do what we've done, presumably at his request.

It's hard to imagine being persuaded, solely by this logical consideration, to abandon all power to form opinions about what would have been wanted by a spouse, a child, a parent, or a friend. But even the slightest awareness of the logical problem may remind us that the comatose patient is only supposed to be making a request—is expressing his desire only in the subjunctive mood. And that, in turn, may remind us that there is a gap between the person we imagine asking us to end his life and the person whose life actually gets ended. One is the man as he was or should be, and the other is the man as he is.

This gap is what drove Mrs. Finn to her intuitively correct use of the subjunctive. Part of the problem, of course, is that to use the indicative "He doesn't want to live like that" would be to raise unanswerable questions about how she could possibly know what he now wants. (Some of Finn's friends and family claimed that he had periods of awareness and responsiveness, but except for a much-disputed incident in which he may have said "Hi" to a newly arrived nurse, Finn remained unspeaking from his accident to his death.)

But, even more, to use the indicative would be to declare that Hugh Finn, lying there unmoving on his hospital bed, is a genuine person with conscious, rational desires. And if he's a genuine person in his current condition, then removing his food and water means at the very least a massive failure to care for a helpless man. It may even mean murder in a particularly brutal and drawn-out way. If he is sufficiently conscious, active, and intelligent to decide that he doesn't want to live as a vegetative patient, then he isn't a vegetative patient.

What that leaves is the subjunctive: an imagining of what he would want if he could somehow know that he is in a condition in which he doesn't know his condition. This is not quite as self-contradictory as it sounds, although it's close. All it requires is that we perform a doubling: picturing one Hugh Finn as the physical body lying comatose on the bed, and a second Hugh Finn as a ghostly presence standing beside us looking down at the other Hugh Finn in pity and horror.

This kind of imaginative doubling has results frightening enough even when patients do it to themselves. In March 1996, the Journal of the American Medical Association published a widely noticed study of euthanasia. Led by a Seattle physician, Anthony Back, the study showed that 26 percent of doctors in Washington State were already granting 24 percent of requests for death. Indeed, it was a prepublication copy of this study that led Judge Stephen Reinhardt to call euthanasia a "time-honored" custom in his Ninth Circuit decision, Washington v. Gluckberg, one of the two cases whose review by the Supreme Court later that year prompted "The Philosophers' Brief." (In his introduction to the brief's publication in the New York Review of Books, Ronald Dworkin also made considerable use of Back's study.)

What hardly anyone noticed at the time, however, was that the vast majority of requests came from patients with particular diseases. An earlier hint of this appeared in a 1989 study in the Annals of Internal Medicine showing that "Do Not Resuscitate" orders are posted more often by cancer and AIDS patients than by patients with heart disease, even when they have similarly poor life expectancy and health quality. In the statistics gathered by Back's team, cancer and AIDS patients were joined by sufferers from progressive neurological decay such as Alzheimer's disease in making over two-thirds of requests for death. Neurological disease accounted for 15 percent of the requests; cancer (causing 25 percent of deaths in Washington State) for 41 percent of requests; and AIDS (1.2 percent of deaths) for a staggering 13 percent. By contrast, heart disease (29 percent of deaths) accounted for only 6 percent of requests for death.

Asked to list reasons for their patients' wish to die, the Washington doctors cited severe pain in 35 percent of cases, but loss of control in 77 percent, being a burden in 75 percent, being dependent in 74 percent, loss of dignity in 72 percent, and severe depression in 55 percent. Given that there are other diseases that provide the same objective reasons to fear dependency, indignity, and depression, the preponderance of requests from patients with cancer, aids, and Alzheimer's can derive only from the patients' perception of their condition as uniquely horrifying. The inevitable psychological burdens of dying seem to weigh unbearably on the victims of these diseases, as they join the American public in believing cancer, AIDS, and Alzheimer's to be the worst of all conditions—to be fates worse than death.

This kind of doubling of the self is probably inevitable. Virtually all conscious patients will at some point stand outside themselves, looking down in pity and horror and disgust at their own condition. And in the case of certain diseases strongly feared by the general culture, the socially constructed reaction of that outside self can overwhelm the inward experience of the real patient. Terminally ill sufferers of heart disease aren't any more likely to live, but they are much less likely to want to die.

That fact alone is disturbing enough to demand hesitation in fulfilling a patient's request for death. There is some division of self involved, some hint of a difference between the person asking for death and the person receiving it, that ought to be worrisome.

But at least the conscious and communicating patient has authority to speak for himself, however divided. At least the terminally ill patient can use the indicative to say what he does want rather than the subjunctive to say what he would want. There remains, in the conscious patient, some reality that ties the imagined, self-conscious, outside self to the real body on the bed and keeps that patient from becoming utterly sundered.

When we presume to speak for those unable to speak for themselves, however, the doubling becomes complete. There is no reality tying the man as he really is to our imagination of him as he was or should be, no way to check the accuracy of our memory, no possibility of obtaining the opinion of the patient on the bed. When we rely on the Finn proposition to support the removal of life support, we kill the actual Hugh Finn at the request of the imagined Hugh Finn—and the imagined Hugh Finn is not the one who will suffer eight days of lethal thirst and hunger if we are wrong.

It was fear of something like this that drove the Finn family to oppose Michele Finn's decision to remove her husband's food and water. "Saying that you wouldn't want to live 'like that' is a far cry from saying you'd prefer to be starved to death," Ed Finn wrote in the Washington Post a week after his brother died. "No one wants to live like that, but neither would anyone want to die like that."

Even this complaint, however, doesn't quite reach down to what is really wrong with using the Finn proposition as a reason for death. In part, that's because Ed Finn is offering what will eventually turn into the grounds for the direct killing of unconscious patients: Once we become fully resigned to what is at present asserted to be merely "allowing the comatose to die," the fact that starvation and dehydration make a cruel death will be the fundamental argument for taking the next step and administering an immediately fatal poison. It is not unreasonable to imagine that the euthanasia activists who this year indignantly denied the suffering in Hugh Finn's death will be the ones who next year indignantly proclaim it.

But in larger part, Ed Finn's complaint is insufficient because it still accepts the possibility of describing what would be wanted by those unable to speak for themselves. When we double Hugh Finn—when we imagine him beside us, looking down upon himself with the same pity and horror we feel—we create a phantasm. That phantasm may have its origin simply in our imagining the unconscious patient to be able to perform the same doubling of self that conscious patients can perform. But it ought to be a matter of suspicion that the imagined Hugh Finn turns out to have the same reaction as we who are imagining him.

This is, in fact, how the Finn proposition transforms an impulse of pity into a demand for death. Faced with a person presumed to be unfeeling, we project upon him our own feelings.

Among the activists fighting the euthanasia movement, there are those who think that this transformation derives ultimately from selfishness: Confronted by an object of care, we pretend that the object itself begs us to stop caring. There are even some who believe that it ultimately derives from a sick, death-loving culture that cannot suffer life under any difficulties.

But there is a way to reject the Finn proposition without imputing selfish motives and a sick psyche to the good, ordinary people tempted to speak for their comatose friends and relatives. The claim of what Hugh Finn would want comes not from Hugh Finn. It comes from us, from a transfer of our emotions onto the man who caused us to have them. This is not feeling for someone, but feeling for someone: a replacement of the self for others.

It is, in fact, something we ought always to beware: the having of another person's reactions for him. Sympathy is typically defined as what we feel while viewing other people's suffering condition, while empathy is instead the attempt to feel their emotions. When we hear "He wouldn't want to live like that" offered as an argument for withdrawing food and water from a patient, we hear the results of substituting our own feelings for those of someone else: pity changed into the pitiless, an object of care converted into an object of death, sympathy remade into murderous and usurping empathy.

J. Bottum is Books & Arts editor of the Weekly Standard.

Most ReadMost Shared